The Organisation of Insecurity

Security, Power and the Systems that Produce Them

Security is not simply the protection of what exists. It is the organisation of relationships that determine what can continue to exist.

Every society depends upon arrangements it cannot completely control. Energy must arrive, information must remain intelligible, food must be distributed, and institutions must retain sufficient legitimacy to function. None of these activities occurs independently. Each changes the circumstances in which the others must continue. Security concerns the persistence of these relationships under conditions that are themselves changing. It is not something a society possesses. It is something a society continually produces, sometimes through activities that also generate new vulnerabilities.

Governments divide security into defence, intelligence, economics, technology, infrastructure and diplomacy. These divisions are administratively necessary, but their boundaries do not determine where consequences travel. An interruption to electricity becomes a communications problem, a medical problem, an industrial problem and eventually a political problem. The significance of a disturbance lies in the dependencies through which its consequences propagate. A small interruption at a critical point can reorganise an entire system. An enormous expenditure elsewhere can accomplish remarkably little. Budgets, unfortunately, remain easier to measure than relationships.

The difficulty is not that institutions lack a sufficiently comprehensive view of the whole. No such view exists. Every observer operates within the circumstances being observed, using information shaped by particular responsibilities, histories and limitations. A security organisation cannot escape these conditions through more elaborate analysis. It can, however, organise differences between perspectives into a more informative account of what is happening. Engineers identify technical dependencies, economists recognise resource constraints, intelligence analysts examine intentions, and communities experience consequences that institutional reporting often fails to register. The objective is not perfect visibility. It is to make the limitations of each perspective consequential to the decisions being made.

This requires boundaries, but boundaries that remain open to revision. A useful assessment begins with a particular decision, identifies the relationships necessary for its intended outcome, and follows the most consequential dependencies and delays. It asks where effects might propagate, who would recognise them, and what evidence would indicate that the original understanding was inadequate. Analysis cannot continue indefinitely. Its scope must reflect the significance of the decision, the available time and the consequences of being wrong. The purpose of examining complexity is not to reproduce the entire world in a briefing document. That would be an ambitious way of ensuring nothing happens.

Security competition demonstrates why these distinctions matter. One state expands its military capabilities to reduce vulnerability. Others interpret that development through their own circumstances and respond. Their responses alter the conditions confronting the first. Defensive intentions do not guarantee defensive consequences. Opposing positions become mutually sustaining through the expectations and activities they generate. Yet not every threat arises from misunderstanding or reciprocal insecurity. Some actors deliberately exploit vulnerabilities, seek domination or pursue destruction regardless of the wider consequences. Recognising systemic interdependence does not require assuming shared interests, reciprocal restraint or benign intentions. Defence remains necessary where threats are deliberate. The additional responsibility is to understand how defensive action changes the circumstances in which those threats subsequently develop.

These processes also have rhythms. Political decisions operate through electoral cycles, military procurement through years of planning, financial transactions through fractions of seconds, and infrastructure through investments that persist for generations. Their interactions involve delays, synchronisation, interference and accumulated consequences. An anticipated threat changes investment before additional capability exists. That investment changes expectations elsewhere, while the eventual capability arrives in circumstances already transformed by the responses it provoked. Timing is not secondary to causation. It is part of the organisation through which consequences arise.

This presents an operational difficulty. Careful analysis requires time, while immediate threats sometimes demand action before their wider implications can be understood. The answer is not to make every tactical decision dependent upon prolonged deliberation. It is to organise learning across different timescales. Strategic analysis identifies dependencies and possible failure patterns before emergencies occur. Operational planning establishes contingencies and limits. Tactical decisions proceed within those preparations, using the information available at the time. Subsequent outcomes revise the assumptions upon which future preparations depend. Reflection and action become different phases of the same continuing activity, rather than competing institutional obligations.

Intervention requires equal caution. A modest change can have disproportionate consequences, but those consequences are not necessarily beneficial or predictable. Interdependent systems do not offer convenient controls through which selected outcomes can be guaranteed. Changes should be proportionate to what is understood, with attention to reversibility, observable consequences and the capacity to stop or adapt when expectations fail. Where possible, limited trials can reveal dependencies before commitments become difficult to reverse. Where immediate action is unavoidable, uncertainty must remain visible rather than being concealed behind the appearance of precision. Learning depends upon retaining the differences between what was expected and what actually occurred.

Technology intensifies these difficulties. Artificial intelligence, autonomous systems and integrated surveillance accelerate information processing, but acceleration does not guarantee understanding. Automated systems inherit the priorities, distinctions and omissions through which they are constructed. When their outputs influence subsequent observations, errors can become recurrent features of the environment. Faster decisions can compress opportunities for reconsideration while extending the consequences of misunderstanding. The relevant question is not simply whether a technology performs its assigned function, but how its operation changes the relationships upon which that function depends.

Resilience raises another problem. Recovering established functions is not necessarily equivalent to improving security. Institutions can survive repeated disturbances while losing the capacity to fulfil their purposes. An economy can maintain production while degrading the resources supporting it. A security organisation can become increasingly effective at responding to threats while helping reproduce the circumstances that make those threats probable. Persistence alone is not evidence of success. What persists, for whom, at whose expense, and with what consequences for future possibilities must remain part of the assessment.

Institutional resistance to change is not simply a failure of understanding. Budgets, professional identities, procurement arrangements, political authority and established responsibilities create material interests in continuity. Some participants benefit from existing arrangements, even when those arrangements impose costs elsewhere. Asking institutions to recognise their own contribution to insecurity is insufficient if doing so threatens the resources or authority upon which their continuation depends. A workable response must alter the conditions under which institutional decisions are rewarded, reviewed and reproduced.

This need not begin with wholesale reorganisation. Existing bureaucracies already contain audits, procurement reviews, exercises, incident investigations, risk registers and performance assessments. These provide opportunities to examine consequences beyond the boundaries of individual departments. A procurement review can consider dependencies introduced elsewhere. An exercise can test how apparently successful decisions generate difficulties for other participants. An incident investigation can examine the organisational conditions that made failure probable rather than merely identifying the last person involved. Independent review, protected dissent and shared accountability can make inconvenient information harder to exclude. Such mechanisms are imperfect and vulnerable to becoming rituals themselves. Their value depends upon whether findings can actually change subsequent decisions.

The practical objective is to make institutional learning part of ordinary operation rather than an exceptional response to catastrophe. This requires observable commitments: identifying consequential assumptions, recording expected outcomes, revisiting decisions when evidence changes, and ensuring that responsibilities extend to effects beyond immediate administrative boundaries. None of these measures requires an institution to understand the entire system. They require it to remain responsive to the consequences of its own participation. The distinction is fundamental. Reflexivity is not an achievement of complete self-knowledge. It is an organised capacity to recognise and revise the conditions through which one’s activities produce further consequences.

Strategic advantage must be understood accordingly. It is not an intrinsic property of a weapon, institution, technology or state. It exists through differences within particular relationships. A capability becomes advantageous because of what others possess, anticipate, require or cannot accomplish. Change those relationships and the advantage changes with them. Technological superiority can introduce dependencies upon fragile infrastructure. Political dominance can diminish the legitimacy required to exercise authority. Successful coercion can reorganise opposition into more persistent forms. Power changes the environment in which its own effectiveness must subsequently be judged.

Security ultimately concerns the conditions under which collective life remains possible. Defence, preparedness and technological development remain necessary, but their effectiveness cannot be assessed independently of the relationships they reorganise. The central strategic question is not simply how to become more capable of confronting threats. It is how our activities participate in producing the circumstances from which threats, vulnerabilities and responses emerge.

A reflexively aware security organisation does not need to wait for a systemic crisis before changing. It needs recurring opportunities to discover where its expectations fail, sufficient institutional freedom to act upon those discoveries, and accountability for consequences that extend beyond its immediate responsibilities. The process begins within existing arrangements, using the differences between their intended functions and their actual effects as material for revision.

The most consequential security failure is not always the inability to recognise a threat. It is the inability to recognise ourselves in the processes that keep producing it.